Unrecognized diabetes in critically ill COVID-19 patients

Dear Editor, Since the first discovery of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and description of the coronavirus disease 2019 (COVID-19), a pandemic has evolved. Due to winter tourism, Tyrol, a federal province of Austria with 750,000 inhabitants, has emerged as an epicenter in Austria being faced with a surge of critically ill COVID-19 patients reaching its peak on April 8, 2020. We retrospectively analyzed the incidence of diabetes in all critically ill patients admitted to the four dedicated COVID-19 intensive care units (ICU) at the University Hospital in Innsbruck, Tyrol, Austria, which covers 180,000 inhabitants as primary hospital and also functions as a tertiary referral center for the whole region of Tyrol. Patients were included in the analysis if they were 18 years of age or older, had confirmed COVID-19, and were admitted to an intensive care unit from March 11 to April 29, 2020. COVID-19 was confirmed by reverse-transcriptasepolymerase-chain-reaction assays of nasopharyngeal swab specimens. Data were abstracted manually from electronic and paper-based health records. Glycated hemoglobin (HbA1c) was measured on admission by high-performance liquid chromatography (HPLC-UV/ VIS). Of 47 COVID-19 patients admitted to our ICUs, HbA1c was measured in 44, which were included in the analysis (Table 1). The median age of patients was 61.5 (IQR 53.0–68.0). Thirty-five (80%) patients required invasive mechanical ventilation (IMV). Additionally, 4 patients (9%) required veno-venous extracorporeal membrane oxygenation (vvECMO). At the time of writing this article, 11 patients (25%) have died in the hospital, 25 (56.8%) have been discharged alive from the ICU, 20 patients (45.5%) were discharged alive from the hospital, and 13 patients (29.5%) are still hospitalized. Median HbA1c was 6.5% (IQR 6.1–6.7%). When categorizing patients according to HbA1c [1], 24 (54.5%) were considered to have diabetes mellitus (HbA1c ≥ 6.5%), 16 (36.3%) were considered to have prediabetes (HbA1c ≥ 5.7% < 6.5%), and only 4 (9%) had no diabetes (HbA1c < 5.7%). Interestingly, only 7 (15.9%) patients showed a medical history of diabetes mellitus. Five (11.4%) patients had previously been treated with antidiabetic medication, and no patient had required insulin prior to hospitalization. Patients with increased HbA1c levels developed higher maximum CRP and IL-6 levels during their ICU stay. There was a trend to higher in-hospital mortality with increasing HbA1c. The median body mass index (BMI) was 29.4 kg/m (IQR 26.2–32.7), which is slightly higher than a previously studied sample of critically ill patients in Austria [2], with a median BMI of 26 kg/m. BMI did not differ significantly between diabetic and non-diabetic patients (Fig. 1). In conclusion, 85% of COVID-19 treated in our intensive care units had prediabetes and diabetes which appear to be predisposing factors for severe manifestations of COVID-19, potentially impairing outcome. This is in line with previous observations from the first SARS-CoV epidemic [3].


Dear Editor,
Since the first discovery of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and description of the coronavirus disease 2019 (COVID-19), a pandemic has evolved. Due to winter tourism, Tyrol, a federal province of Austria with 750,000 inhabitants, has emerged as an epicenter in Austria being faced with a surge of critically ill COVID-19 patients reaching its peak on April 8, 2020.
We retrospectively analyzed the incidence of diabetes in all critically ill patients admitted to the four dedicated COVID-19 intensive care units (ICU) at the University Hospital in Innsbruck, Tyrol, Austria, which covers 180,000 inhabitants as primary hospital and also functions as a tertiary referral center for the whole region of Tyrol. Patients were included in the analysis if they were 18 years of age or older, had confirmed COVID-19, and were admitted to an intensive care unit from March 11 to April 29, 2020. COVID-19 was confirmed by reverse-transcriptasepolymerase-chain-reaction assays of nasopharyngeal swab specimens. Data were abstracted manually from electronic and paper-based health records. Glycated hemoglobin (HbA1c) was measured on admission by high-performance liquid chromatography (HPLC-UV/ VIS).
The median body mass index (BMI) was 29.4 kg/m 2 (IQR 26.2-32.7), which is slightly higher than a previously studied sample of critically ill patients in Austria [2], with a median BMI of 26 kg/m 2 . BMI did not differ significantly between diabetic and non-diabetic patients (Fig. 1).
In conclusion, 85% of COVID-19 treated in our intensive care units had prediabetes and diabetes which appear to be predisposing factors for severe manifestations of COVID-19, potentially impairing outcome. This is in line with previous observations from the first SARS-CoV epidemic [3].
Hyperglycemia may alter the response of the innate immune system through several mechanisms. It may induce Toll-like receptor expression and inhibit neutrophil function, decrease vascular dilation, and increase permeability [4]. Furthermore, it can cause direct glycosylation of proteins, thereby altering the structure of complement, and may cause a cytokine storm [4,5]. Recent data demonstrating viral particles in endothelial cells of several organs suggest "endotheliitis" as a possible mechanism of organ dysfunction leading to critical illness in COVID-19 patients which may be aggravated by endothelial  Abbreviations: IQR interquartile range, BMI body mass index, HbA1c glycated hemoglobin, CRP C-reactive protein, IL-6 interleukin-6, COPD chronic obstructive pulmonary disease, SARS-CoV-2 severe acute respiratory syndrome coronavirus 2 *If specified in the patients' health records dysfunction associated with prediabetes and diabetes [6]. More pronounced peak levels of inflammation observed in our patients with abnormal HbA1c may support such an assumption. In conclusion, we recommend routine measurement of HbA1c in hospitalized COVID-19 patients for additional risk stratification, because most patients of our cohort were previously not diagnosed with having impaired glucose tolerance.